The Arabidopsis COG1 gene encodes a Dof domain transcription factor and negatively regulates phytochrome signaling

Transcription

1 The Plant Journal (2003) 34, 161±171 The Arabidopsis COG1 gene encodes a Dof domain transcription factor and negatively regulates phytochrome signaling Don Ha Park y, Pyung Ok Lim y, Jeong Sik Kim, Dae Shik Cho, Sung Hyun Hong and Hong Gil Nam Division of Molecular Life Sciences, Pohang University of Science and Technology, Pohang, Kyungbuk , Korea Received 30 September 2002; revised 30 December 2002; accepted 9 January For correspondence (fax ; nam@bric.postech.ac.kr). y These two authors made equal contributions. Summary Light is a critical environmental factor that in uences almost all developmental aspects of plants, including seed germination, seedling morphogenesis, and transition to reproductive growth. Plants have therefore developed an intricate network of mechanisms to perceive and process environmental light information. To further characterize the molecular basis of light-signaling processes in plants, we screened an activation tagging pool of Arabidopsis for altered photoresponses. A dominant mutation, cog1-d, attenuated various red (R) and far-red (FR) light-dependent photoresponses. The mutation was caused by overexpression of a gene encoding a member of the Dof family of transcription factors. The photoresponses in Arabidopsis were inversely correlated with the expression levels of COG1 mrna. When the COG1 gene was overexpressed in transgenic plants, the plants exhibited hyposensitive responses to R and FR light in a manner inversely dependent on COG1 mrna levels. On the other hand, transgenic lines expressing antisense COG1 were hypersensitive to R and FR light. Expression of the COG1 gene is light inducible and requires phytochrome A (phya) for FR light-induced expression and phytochrome B (phyb) for R light-induced expression. Thus, the COG1 gene functions as a negative regulator in both the phya- and phyb-signaling pathways. We suggest that these phytochromes positively regulate the expression of COG1, a negative regulator, as a mechanism for ne tuning the light-signaling pathway. Keywords: Dof transcription factor, light signaling, phytochrome, activation tagging. Introduction Plants monitor various aspects of light, such as wavelength, intensity, direction, and period, and incorporate this information into their developmental programs (Neff et al., 2000). Light signals are rst perceived by plants' photoreceptors. Phytochromes are the photoreceptors that have been best characterized at the biochemical, molecular, and physiological levels. Phytochromes can be reversibly photo-converted between the red-light-absorbing Pr form and the far-red-absorbing Pfr form. This photo-convertibility enables phytochromes to sense different ratios of red to far-red light and to function as photo-reversible molecular switches (Furuya, 1993; Neff et al., 2000; Quail et al., 1995). The Arabidopsis genome encodes ve members of the phytochrome gene family, PHYA to PHYE (Quail et al., 1995). The PHYB to PHYE genes encode lightstable phytochromes. In contrast, the PHYA gene encodes a light-labile phytochrome that accumulates at high levels in darkness, but rapidly degrades upon exposure to light. The light-stable and light-labile phytochromes also share overlapping functions in controlling light-mediated plant development (Quail et al., 1995; Smith, 2000; Whitelam and Devlin, 1997). Phytochrome A (phya) is the primary, if not the only, photoreceptor responsible for seedling de-etiolation, including inhibition of cotyledon opening and hypocotyl elongation in continuous FR (FRc) light (Nagatani et al., 1993; Parks and Quail, 1993; Whitelam et al., 1993). PhyA is also responsible for FRc-induced anthocyanin accumulation and FR-preconditioned blocking of greening (Barnes et al., 1996; Kunkel et al., 1996). Although many phya responses are observed in FR light, phya also has distinctive functions in R and other types of light through `very low uence' ß 2003 Blackwell Publishing Ltd 161

2 162 Don Ha Park et al. responses to control seed germination, R-modulated phototropism, and other functions (Botto et al., 1996; Parks et al., 1996; Reed et al., 1993; Shinomura et al., 1996). In contrast, phyb is the primary photoreceptor that mediates seedling de-etiolation responses in continuous R (Rc) light (Nagatani et al., 1991; Reed et al., 1993; Somers et al., 1991). PhyB also plays a role in other R light-mediated responses, such as inhibition of stem and petiole elongation, delay of owering, and the end-of-day (EOD) far-red response (Nagatani et al., 1991; Neff and Van Volkenburgh, 1994; Reed et al., 1993). Many of the phyb responses exhibit classical R/FR photo-reversibility. Because they have both distinct and overlapping functions, it has been suggested that phya and phyb may control downstream responses through three major signaling branches (Neff et al., 2000; Quail, 2002). A phya-speci c signaling branch is de ned by components, such as FHY1, FHY3, FIN2, FIN219, SPA1, FAR1, PAT2, EID2, HFR1, and LAF1 (Ballesteros et al., 2001; Bolle et al., 2000; Fairchild et al., 2000; Hoecker et al., 1998, 1999; Hsieh et al., 2000; Hudson et al., 1999; Soh et al., 1998, 2000; Whitelam et al., 1993). The phyb-speci c branch is de ned by components, such as PEF2, PEF3, POC1, PKS1, RED1, and SRL1 (Ahmad and Cashmore, 1996; Halliday et al., 1999; Huq et al., 2000; Wagner et al., 1997). In contrast, PIF3, PEF1, PSI2, and NDPK2 function at a branch common for both the phyaand phyb-signaling pathways (Ahmad and Cashmore, 1996; Choi et al., 1999; Fankhauser et al., 1999; Genoud et al., 1998; Ni et al., 1998). The various components of the phytochrome-signaling pathway can be classi ed as mediating either negative or positive regulation. Molecular and genetic approaches have revealed several negative components in the phytochrome-signaling pathways (Quail, 2002). SPA1, EID1, and SUB1 function as negative regulators in phya signaling (Dieterle et al., 2001; Guo et al., 2001; Hoecker et al., 1999). In the phyb-signaling pathway, SRL1, PKS1, and PIF4 act as negative components (Huq and Quail, 2002; Huq et al., 2000). PSI2 has been proposed to act as a negative component in both the phya- and the phyb-signaling pathways (Fankhauser et al., 1999; Genoud et al., 1998). Interestingly, the majority of phytochrome-signaling molecules identi- ed thus far appear to function as positive components. Part of the reason for this may be that most of the earlier genetic screening efforts to identify phytochrome-signaling components have focused on isolating loss-of-function mutations that exhibit reduced sensitivity to light. We have identi ed components that regulate lightmediated development in Arabidopsis by isolating mutations speci c to the phya pathway (Soh et al., 1998) and suppressor mutants of hy2, a phytochrome-de cient mutant (Kim et al., 1996, 1998). In a further attempt to isolate novel components in light signaling, we screened a T-DNA activation tagging pool (Weigel et al., 2000) because T-DNA activation tagging generates gain-of-function mutations and often reveals the function of genes that are not revealed in loss-of-function mutations. Through this screening, we identi ed a dominant mutant, cog1-d, that showed defects in both phya- and phyb-mediated light responses. The mutation was caused by overexpression of a gene encoding a member of the plant-speci c Dof (DNA binding with one finger) domain-containing a family of transcription factors. Through characterization of the mutant phenotypes, we suggest that COG1 functions as a negative regulatory component within both the phya- and phyb-signaling pathways. Furthermore, expression analysis of the COG1 gene suggests that it may act through a novel mechanism to control the ow of light signaling to downstream responses. Results Identification of a dominant mutant cog1-d from an activation tagging population of Arabidopsis The activation tagging pool was generated using the pski015vector, as described previously (Weigel et al., 2000). Approximately 7000 individual T 2 families were screened for a long hypocotyl phenotype in white light. Through this screening, we found a single T 2 family that exhibited segregation of the long hypocotyl phenotype (data not shown). The segregation ratio of the mutant phenotype in the T 2 generation was approximately 3 : 1 (mutant : wild type). This segregation ratio suggested that the phenotype was caused by a single dominant allele and that activation of a gene most likely mediated the phenotype in the mutant plant. The seedling with the long hypocotyl phenotype later developed leaves with long petioles and blades reduced in size (data not shown). Inhibition of petiole elongation and expansion of leaf blades are controlled in part by light, and are thereby frequently observed in photomorphogenic mutations (Devlin et al., 1998; Huq et al., 2000). These phenotypes are also consistent with the hypothesis that this mutation has a defect in light responses. The mutation was named cog1-d because the gene functions like a cogwheel in the light-signaling pathway (see below). An Arabidopsis line homozygous for the mutant phenotype was isolated in the T 3 generation and was examined for hypocotyl growth in red (R), far-red (FR), and blue (B) light (Figure 1a). When grown in darkness, cog1-d seedlings exhibited a hypocotyl length indistinguishable from that of wild-type seedlings. In contrast, when grown in various types of light, the mutant seedlings developed hypocotyls longer than those of wild-type seedlings and exhibited a defect in light-mediated inhibition of hypocotyl growth. Although the long hypocotyl phenotype was

3 Role of COG1 in phytochrome signaling 163 Figure 1. Altered photoresponses of cog1-d mutant. (a) Seedling phenotypes. The seedlings were grown on MS medium for 5days in 60 mmol m 2 sec 1 of red light (R), 4 mmol m 2 sec 1 of far-red light (FR), 6 mmol m 2 sec 1 of blue light (B), or in darkness (D). Shown in each set, from left to right, are Col-0, cog1-d, phya-211, phyb- 9, Ler, and hy4. Scale bar, 5mm. (b) Light-regulated expression of the CAB2 gene. Wild type (wt), cog1-d, phya-211, and phyb-9 seedlings were grown for 5days in darkness and then exposed to either R light (150 mmol m 2 sec 1 ) or FR light (30 mmol m 2 sec 1 ). After 0, 0.5, or 3 h of light exposure, total RNA was isolated. Samples containing 5 mg of total RNA were subjected to RNA-gel blot analysis for CAB2, as described in Experimental procedures. The 18S rdna probe was used as a loading control. (c) Total cellular RNA was extracted from seedlings of wild-type (wt), COG1 antisense line (cogas6), cog1-d mutant, and COG1 overexpressing line (cogox1) plant seedlings that were grown in white light (1.1 W m 2 ) for 7 days. Twenty micrograms of total RNA was loaded in each lane and subjected to RNA-gel blot analysis for COG1, as described in Experimental procedures. The 18S rdna probe was used as a control. The numbers indicate the induction level of COG1 calculated from the integrated density of the bands. observed under all types of light that were tested, the phenotype was more prominent in R and FR light than in B light (Figure 1a). We therefore focused our subsequent analyses of the cog1-d mutation on phytochrome responses mediated by R and FR light. Figure 3. Subcellular localization of COG1-GFP protein. A construct encoding green uorescent protein (GFP) was fused to the fulllength COG1 cdna sequence. Either the GFP:COG1 construct (a and c) or the vector control (GFP alone, b and d) were introduced into onion epidermal cells by particle bombardment. After 18 h of incubation, samples were analyzed for GFP uorescence (a and b) or stained with 4 0,6 0 -diamidino-2- phenylindole (DAPI) (c and d). The cog1-d mutation confers lesions in other phytochrome-mediated responses In addition to inhibition of hypocotyl elongation, it is also known that the cotyledon opening response is controlled by phya in FR light and by phyb in R light (Reed et al., 1993; Yanovsky et al., 1997). To further determine whether R and FR responses are defective in the cog1-d mutant, the cotyledon opening response was measured in various uences of R and FR lights (Table 1). In R light, the cotyledon opening response of the mutant was greatly reduced in all uences examined (0.7±50 mmol m 2 sec 1 ). In FR light, prominent differences in the cotyledon opening response between mutant and wild-type seedlings were observed at a low uence rate (0.2 mmol m 2 sec 1 ), and somewhat smaller differences were observed at higher uence rates. These results show that the cog1-d mutation confers reduced sensitivity of cotyledon opening in response to R and FR light. In addition to morphogenesis, gene expression is also regulated in response to FR and R lights (Ma et al., 2001; Tepperman et al., 2001). We next determined whether the cog1-d mutation also affects expression of a light-inducible gene. For this purpose, we analyzed mrna levels of the gene encoding the chlorophyll a/b-binding protein (CAB2) upon transferring seedlings from darkness to R or FR light

4 164 Don Ha Park et al. Table 1 Red and far-red light-induced cotyledon opening in Col-O and cog1-d mutant Col-O Angle cog1-d Red (mmol m 2 sec 1 ) (6.0) 19 (2.2) 2 97 (4.9) 25(2.1) (4.0) 34 (4.9) (3.6) 49 (5.4) (6.2) 73 (4.1) Far-red (mmol m 2 sec 1 ) (7.3) 12 (3.2) (5.1) 27 (8.1) (8.7) 48 (11.6) (6.6) 117 (7.2) (7.0) 176 (7.4) Plants were grown for 4 days under various fluences of red and far-red light. The angles between the two cotyledons were measured. Shown are the average angle values of 25samples in each measurement. Values in parentheses are SE. (Figure 1b). The light-induced increase in CAB2 mrna levels in the mutant was reduced both in R and FR light, as compared to that of wild-type seedlings. In both R and FR light, the CAB2 mrna levels almost reached those of wild type after prolonged exposure. The decrease in CAB2 mrna levels in the cog1-d mutant was not as pronounced as that observed in the phya or phyb mutants, which is consistent with the hypocotyl length phenotypes of these mutants. A gene encoding a Dof transcription factor is overexpressed in the cog1-d mutant The cog1-d mutation was isolated from an activation tagging pool and the mutant phenotype segregated as a dominant allele. Thus, the mutation was unlikely to result from an insertional knock-out of a gene, but rather from activation of a gene. To further characterize the cog1-d mutant, we cloned the genes that are activated in this activation-tagging mutation. Genomic DNA-gel blot analysis revealed that the mutation contained a single T-DNA insertion in the genome (data not shown). The genomic DNA fragment anking the right border of the T-DNA insertion was isolated by plasmid rescue, utilizing the single EcoRI site in the T-DNA. The sequence of the genomic DNA anking the T-DNA was then compared to the entire genomic sequence of Arabidopsis. T-DNA activation tagging often results in overexpression of the nearest gene from the CaMV35S enhancers to induce the mutant phenotype (Kakimoto, 1996; Kardailsky et al., 1999; Weigel et al., 2000). An open reading frame (ORF) closest to the right border of the T-DNA was 528 bp long and encoded 175 amino acids, as shown in Figure 2. RNA-gel blot analysis revealed that expression of the ORF was indeed activated in the mutant plants (Figure 1c). This suggested that overexpression of the gene for the ORF is responsible for the cog1-d mutation phenotype. The predicted amino acid sequence of the gene showed considerable similarity to a group of Dof domain-containing proteins. The Dof domain is found in a family of DNA-binding transcription factors that are unique to plants (Yanagisawa, 1995, 2002; Yanagisawa and Schmidt, 1999). The domain includes a Figure 2. COG1 encodes a Dof transcription factor protein. (a) The deduced amino acid sequence of the COG1 gene. The highly conserved Dof DNA-binding domain is underlined, and a predicted nuclear localization signal (PVKRLRC) is shown in bold. The predicted nuclear localization signal was identi ed by PSORT II software ( (b) Alignment of the Dof domains from various Dof proteins in Arabidopsis using the CLUSTAL X software (Thompson et al., 1997). Conserved and similar amino acid residues are highlighted in black and gray shading, respectively. COG1 (At1g29160), ADOF1 (At1g51700), ADOF2 (At3g21270), DAG1/BBFa (At3g61850), DAG2 (At2g46590), OBP1 (At3g50410), OBP2 (At1g07640), OBP3 (At3g55370), and OBP4 (At5g60850).

5 Role of COG1 in phytochrome signaling 165 highly conserved 52-amino acid stretch with a single zinc nger that mediates DNA binding (Yanagisawa, 1995). The predicted COG1 peptide sequence contains the highly conserved 52-amino acid Dof domain (Figure 2b). To determine whether overexpression of the COG1 gene mediates the mutant phenotypes observed, we generated transgenic plants overexpressing COG1 driven by the CaMV35S promoter. Among the 65 individual transgenic families, 46 families exhibited the long hypocotyl phenotype when grown in light (data not shown). These ndings con rm that the mutant phenotype was caused by overexpression of the ORF (COG1) encoding the Dof transcription factor. COG1 is localized to the nucleus The amino acid sequence of COG1 contains a putative nuclear localization signal (PVKRLRC) near the carboxyl terminus (Figure 2a). Thus, we used a 35S:green uorescent protein (GFP)-COG1 fusion construct to determine the subcellular localization of COG1 in onion epidermal cells. As shown in Figure 3, 35S:GFP-COG1 was most strongly detected in the nucleus, whereas a 35S:GFP construct did not exhibit any distinctive localization pattern. These ndings suggest that COG1 encodes a nuclear localized Dof transcription factor. The expression level of COG1 is inversely correlated with the degree of light-induced inhibition of hypocotyl growth To investigate the role of COG1 in phytochrome-mediated hypocotyl growth inhibition, we examined hypocotyl growth in response to a range of R and FR uence rates in wild-type, cog1-d mutant, COG1-overexpressing transgenic (cogox1), and COG1 antisense (cogas6) seedlings. The transgenic line, cogox1, was chosen from the transgenic lines that express COG1 gene under the CaMV 35S promoter (see above). The antisense transgenic line was generated with 180 bp of the 3 0 portion of the proteincoding region of COG1 and 420 bp of the 3 0 untranslated region. This fragment was expressed in the antisense orientation under the control of the dual CaMV35S promoter in the transgenic plants. As shown in Figure 1(c), the expression level of the COG1 mrna was higher in the cogox1 line than in wild-type plants. The COG1 mrna was more abundant in the cog1-d mutant line than in the cogox1 line. In the cogas6 line, the COG1 transcript level was much lower than that of wild-type seedlings. Fluence rate±response curves con rmed that both Rc and FRc light-mediated inhibition of hypocotyl growth was less sensitive in cog1-d seedlings than in wild-type seedlings at all uence rates that were tested (Figure 4a,b). Figure 4. Fluence rate-dependent inhibition of hypocotyl elongation in cog1-d mutant under red and far-red light. Seedlings were grown for 5days in various uence rates of continuous (a) R or (b) FR light, as indicated. (a) Hypocotyl lengths (in mm) of wild-type (wt), cog1-d, cogox1, cogas6, and phyb-9 seedlings were measured in response to growth under various uence rates of continuous R light. (b) Hypocotyl lengths (in mm) of wild-type (wt), cog1-d, cogox1, cogas6, and phya-211 seedlings were measured in response to growth under various uence rates of continuous FR light. The hypocotyl growth response of cogox1 seedlings was intermediate between those of wild-type and cog1-d seedlings. In contrast, hypocotyl growth was more severely inhibited in the antisense seedlings than in wild type in both Rc and FRc light, at all uence rates that were tested.

6 166 Don Ha Park et al. Thus, Rc- and FRc-induced hypocotyl growth inhibition is inversely correlated with the expression levels of the COG1 gene. COG1 controls phyb-mediated end-of-day (EOD)-FR response While hypocotyl growth inhibition is a typical response to high irradiance, Arabidopsis seedlings exhibit various other light responses that are mediated by phytochromes. To further con rm the role of COG1 in phytochromemediated light responses, we examined other easily quanti able light responses in wild-type, cog1-d mutant, cogox1, and cogas6 seedlings. We rst measured the EOD-FR response, which is a low light phyb-mediated response. When wild-type seedlings are grown in a light± dark cycle and treated with a pulse of FR light at the end of the light period, hypocotyl growth is greater than in seedlings that do not receive the FR pulse (Figure 5a; Robson et al., 1993). This EOD-FR pulse reduces the relative amount of PfrB and results in elongated hypocotyls (Devlin et al., 1996). The EOD-FR effect is nulli ed by a subsequent pulse of R light. The EOD-FR response was markedly reduced in seedlings lacking phyb (Figure 5a), as phyb is the major component involved in the EOD-FR response. The cog1-d mutant exhibited a greatly reduced EOD-FR response, comparable to that of the phyb-9 mutant. The cogox1 line exhibited a slightly attenuated EOD-FR response that was intermediate between the responses of the wild-type and cog1-d mutant seedlings. In contrast, the antisense line cogas6 showed an enhanced EOD-FR response, as compared to wild type (Figure 5a). These ndings demonstrate that COG1 regulates the phyb-mediated EOD-FR response and that the expression level of the COG1 gene is inversely correlated with the degree of the EOD-FR response. These results are consistent with the different hypocotyl growth responses of the various genotypes to R light. COG1 controls phya-mediated anthocyanin accumulation Accumulation of anthocyanin in Arabidopsis seedlings grown in FRc light is a phya-mediated response (Kunkel et al., 1996; Neff and Chory, 1998). We next asked if COG1 plays a role in this process. In wild-type seedlings, FR light induced accumulation of anthocyanin in a uence ratedependent manner. In contrast, phya-211 mutant seedlings did not exhibit accumulation of anthocyanin in response to FR light (Figure 5b; Neff and Chory, 1998). The cog1-d mutant seedlings accumulated signi cantly lower levels of anthocyanin than did wild-type seedlings. In the two uence rates tested, the levels of anthocyanin in cog1-d mutants were 63 and 67% lower, respectively, than those of the wild type (Figure 5b). Reduced levels of anthocyanin Figure 5. End-of-day (EOD)-FR response and anthocyanin accumulation in cog1-d mutant. (a) Wild-type (wt), cog1-d mutant, cogox1, cogas6, and phyb-9 seedlings were grown under 10 h white light/14 h dark cycles for 4 days. Seedlings were treated with R light (5min; 30 mmol m 2 sec 1 ) and/or FR light (5min; 10 mmol m 2 sec 1 ) at the end of each of the 10 h light cycle, as indicated. Hypocotyl lengths are expressed as percentage of the average hypocotyl length in seedlings that had received no EOD light treatment. EOD-R treatments are represented by white bars, EOD-FR treatments are represented by black bars, and EOD-FR/R treatments are represented by gray bars. The average hypocotyl lengths of control plates were 6.4 mm for wt, 8.4 mm for cog1-d, 6.9 mm for cogox1, 5.1 mm for cogas6, and 9.4 mm for phyb-9 seedlings, respectively. The error bars indicate standard deviations from three independent measurements. Each measurement was performed with approximately 50 seedlings. (b) Average anthocyanin contents of wild-type (wt), cog1-d mutant, cogox1, cogas6, and phya-211 seedlings grown in FRc (10 mmol m 2 sec 1 and 20 mmol m 2 sec 1 ) for 4 days are shown. The error bars indicate standard deviations of three independent measurements. Each measurement was performed with approximately 200 seedlings. induction were also observed in cogox1 seedlings. At the higher uence rate tested, the levels of anthocyanin accumulation in cogox1 seedlings were intermediate between those of wild-type and cog1-d seedlings. In contrast, the level of anthocyanin was higher in antisense (cogas6)

7 Role of COG1 in phytochrome signaling 167 Figure 6. Phytochrome-regulated expression of the COG1 gene. Wild-type (wt), phyb-9, and phya-211 seedlings were grown in darkness for 5days and exposed to (a) 150 mmol m 2 sec 1 of R or (b) 30 mmol m 2 sec 1 of FR light for 0, 2, and 4 h. Total RNA was isolated and subjected to COG1 RNA-gel blot analysis as described in Experimental procedures. The 18S rdna probe was used as a loading control. seedlings, as compared to wild-type seedlings. Anthocyanin levels did not differ between the various genotypes when seedlings were grown in darkness. These results demonstrate that COG1 regulates phya-mediated anthocyanin accumulation. Consistent with the results of hypocotyl growth and the EOD-FR response, anthocyanin accumulation is also inversely correlated with the expression levels of the COG1 gene. The COG1 gene is induced by R and FR light We determined whether expression of the COG1 gene itself is regulated by light. When dark-grown wild-type seedlings were exposed to R or FR light, the level of COG1 mrna was elevated within 2 h (Figure 6). When exposed to R light, COG1 mrna levels were slightly decreased after 4 h, but were slightly increased 4 h after exposure to FR light. The induction of COG1 mrna in R and FR light suggests that expression of this gene is under the control of a phytochrome. To further address this issue, we examined COG1 mrna levels in phya and phyb mutant seedlings. Unlike wild-type seedlings, phya and phyb mutant seedlings did not exhibit any detectable increase in COG1 mrna levels in response to exposure to FR and R lights (Figure 6). These ndings suggest that phya and phyb play a role in the light-induced regulation of COG1 gene expression. Discussion COG1 is a critical component in light-dependent responses in Arabidopsis The cog1-d mutant was isolated from an activation tagged mutant pool. This line exhibits hyposensitivity to various photoresponses, including light-mediated inhibition of hypocotyl elongation, cotyledon opening, the EOD-FR response, anthocyanin accumulation, and light-inducible gene expression. The defect in the cog1-d mutation is caused by overexpression of a Dof transcription factor gene. This was con rmed by generating transgenic lines that overexpress the COG1 gene and by demonstrating that the phenotype in the cog1-d mutation was recapitulated in the transgenic lines. The phenotypes in a gain-of-function line could be because of a pleiotropic effect and do not necessarily re ect its role in wild-type plants accurately. The limited conclusions that can be drawn from an overexpressing phenotype can be overcome by examining a loss-of-function mutation. We were unable to isolate a knock-out line for the COG1 gene, perhaps because of its small size. However, we did generate an antisense transgenic line in which the expression of COG1 mrna was signi cantly reduced. In the antisense lines, we observed that several light-dependent responses were reversed, as compared to the cog1-d mutant. This con rmed that the COG1 gene indeed functions as a critical component in light-dependent signaling responses in Arabidopsis. COG1 encodes a Dof transcription factor COG1 was localized to the nucleus and encodes a Dof transcription factor (Figures 2 and 3). The Dof domaincontaining proteins comprise a family of plant-speci c transcription factors that are not found in yeast or animals (Yanagisawa, 1995). To date, the Dof family of proteins has been implicated in light-dependent gene regulation in maize (Yanagisawa and Sheen, 1998), in activation of storage protein genes in maize and barley (Mena et al., 1998; Vicente-Carbajosa et al., 1997), and in regulation of stressrelated genes and seed germination in Arabidopsis (Gualberti et al., 2002; Kang and Singh, 2000). Furthermore, a microarray analysis showed that one of the Dof transcription factors (H-promoter binding factor 2A) is induced by FR light, leading to the suggestion that this Dof factor may be involved in phytochrome signaling (Tepperman et al., 2001). While most of these studies provided indirect evidence for a role of the Dof proteins, our results conclusively show that a member of the Dof family plays an important role in phytochrome signaling in plants. The zinc nger of Dof proteins is distinct from other known zinc ngers in terms of amino acid sequence and the arrangement of cysteine residues that co-ordinate the zinc ions that are required for DNA-binding activity (Yanagisawa, 1995). Analysis of the complete genomic sequence of Arabidopsis indicates that it includes approximately 37 members of the Dof gene family (Papi et al., 2002; Riechmann et al., 2000; Yanagisawa, 2002). COG1, like the other Dof family members, is characterized by a highly

8 168 Don Ha Park et al. conserved stretch of 52 amino acids with a single CX 2 CX 21 CX 2 C zinc nger, as shown in Figure 2. However, the amino acid sequences outside the Dof domain are widely divergent. The divergent sequences in COG1 presumably mediate functions that are unique to COG1. Among the Dof family members present in the Arabidopsis genome, one member showed a striking 75% similarity to COG1 at the amino acid level, even in the regions outside of the Dof domain (TrEMBL accession number: O22967). It will be of considerable interest to determine whether this sequence is functionally related to COG1. COG1 functions as a negative component in both phya-and phyb-signaling pathways The phytochrome signaling network includes both positive and negative regulators (Quail, 2002). Such regulators are classi ed as either phya- or phyb-speci c components, or components that mediate signals from both these phytochromes. Our results show that the cog1-d mutation, the transgenic overexpressing lines, and the antisense lines confer defects in both phyb- and phya-mediated responses. Thus, COG1 can be classi ed as a component involved in both phya and phyb signaling. When the phenotypes of the cog1-d mutant, the overexpressing (cogox1), and the antisense lines (cogas6) were compared with those of wild-type seedlings, a clear inverse relationship between the severity of the defects in light responses and the levels of COG1 mrna was apparent. Thus, COG1 functions as a negative regulator in the phya- and phyb-signaling pathways. Compared to the number of positive components identi ed in the phytochrome pathways (Quail, 2002), relatively few negative components have been identi ed thus far. The present study adds COG1 to the small, but growing list of negative regulatory signaling components in both the phya and phyb pathways. Previous studies showed that several transcription factors function as positive regulators in phytochrome signaling in Arabidopsis. PIF3, PIF4, and HFR1 contain a basic helix-loop-helix (bhlh) motif (Fairchild et al., 2000; Ni et al., 1998; Soh et al., 2000; Spiegelman et al., 2000). CCA1 and LHY1 contain a single MYB domain (Schaffer et al., 1998; Wang and Tobin, 1998). HY5and LAF1 contain a basic leucine zipper motif and an R2R3-MYB transcription factor (Ballesteros et al., 2001; Oyama et al., 1997), respectively. Thus far, PIF4 (a bhlh protein) and ATHB2 (a homeodomain protein) have been characterized as negative regulators in phytochrome signaling. Our report adds a Dof domain-containing transcription factor as a negative component in phytochrome signaling. Thus, it appears that, as in the case of positive regulation, diverse transcription factors with different DNA binding motifs are also involved in the negative regulation of phytochrome signaling. COG1 may provide a fine tuning mechanism in phytochrome signaling While the COG1 functions as a negative regulator in both the phya- and phyb-signaling pathways, its R light-dependent expression requires phyb and its FR light-dependent expression requires phya (Figure 6). Light-induced expression of negative regulatory genes has also been observed in the case of PIF4 and SPA1 (Hoecker et al., 1999; Huq and Quail, 2002). Such induction of both positive and negative regulators would allow precise regulation of phytochrome signaling, desensitizing the phytochrome-signaling pathway and preventing excessive one-way signal ow from phytochromes to downstream responses. Furthermore, the downstream light response observed in our analysis exhibits a quantitative relationship with the expression level of the COG1 gene. These properties would be expected to enable COG1 to further ne-tune phytochrome-signaling pathways. Experimental procedures Plant materials and growth conditions All mutants used in this study were on the Col-0 background except for hy4 (Ler). The phya-211 and phyb-9 mutants were obtained from the Arabidopsis Biological Resource Center (Columbus, OH, USA). Arabidopsis was transformed by the oral dipping method (Clough and Bent, 1998). Activation-tagged transgenic lines were generated as described by Weigel et al. (2000). The light sources used in this experiment have been described by Kim et al. (1996). For measurement of hypocotyl length, seeds were sown on Murashige±Skoog (MS) medium containing 0.8% agar. After cold treatment at 48C for 3 days, the plates were placed in white light (WL, 4 W m 2 ) for 8 h at 238C to facilitate germination. Plates were then transferred to the appropriate light conditions. Plasmid rescue Restriction enzyme digestion was used to rescue the sequence adjacent to the right T-DNA border. One microgram of plant genomic DNA was digested with the EcoRI in a 100 ml reaction. After phenol±chloroform extraction, the sample was ligated overnight at 158C in a total volume of 100 ml. The ligated DNA was precipitated and transformed into Eschericia coli DH5a by electroporation. Plasmid construction and sequence analysis For generation of transgenic lines that overexpress the COG1 gene, the entire open reading frame was ampli ed by the polymerase chain reaction (PCR) using two oligonucleotides (5 0 -ATTT- CCATGGCGACCCAAGATTCTCAAGGGA-3 0, where the underlined section is an NcoI site, and 5 0 -TCGGGGTGACCTTAACAAGATT- GTCCATCG-3 0, where the underlined section is a BstEII site). The resulting PCR product was digested with NcoI and BstEII and introduced into the NcoI and BstEII sites of the plant transformation vector, pcambia 3301 (Cambia, Canberra, Australia). To generate the antisense line, a DNA fragment containing 180 bp

9 Role of COG1 in phytochrome signaling 169 of the carboxyl-terminal coding region and 420 bp of 3 0 untranslated region was isolated by PCR using two primers (5 0 -ATAA- GAAAGCGGCCGCTACGTACACTTAAACGATGCGAA-3 0, where the underlined section is a NotI site, and 5 0 -CGGAATTCACGTG- TCGGTGGGTTCGCTGAGTT-3 0, where the underlined section is an EcoRI site). The PCR product was digested with NotI and EcoRI and introduced into NotI and EcoRI sites of pnb96 vector. These plasmids were then introduced into Agroacterium tumefaciens AGL1 by electroporation and transformed into A. thaliana (Col-O) by vacuum in ltration. Sequence alignment was performed using the ClustalX (version 1.8; Thompson et al., 1997) running under default parameters, and BIOEDIT software (version 5.0.9; Hall, 1999) was used for further manipulation of the sequences. Measurement of anthocyanin content, cotyledon opening, and the EOD-FR response Measurements of anthocyanin accumulation were performed as described previously (Mancinelli, 1990; Neff and Chory, 1998; Soh et al., 1998). Seedlings were grown on 0.5 Gamborg's B5 (GibcoBRL, USA) medium, 1.5% (w/v) sucrose, and 0.7% (w/v) agar under 11 or 19 mmol m 2 s 1 of FR light for 4 days. The degrees of cotyledon opening were measured as described previously (Neff and Chory, 1998). For measurement of the EOD-FR response, seedlings were grown in a short-day growth chamber for 4 days under a daily cycle of 10 h of white light (1.1 W m 2 ) and 14 h of darkness. Seedlings were then exposed to FR and/or R light treatment (10 mmol m 2 s 1 ) at the end of each day for 5min. RNA-gel blot analysis Total RNA was extracted using the Tri-Reagent kit (Molecular Research Center, Cincinnati, OH, USA), according to the manufacturer's recommended conditions. RNA was then subjected to formaldehyde-agarose gel electrophoresis and RNA-gel blotting, using standard methods. The COG1 DNA fragment for antisense transgenic plant generation was used for generating a probe. The CAB2 DNA probe was obtained from Dr J. Chory (The Salk Institute, La Jolla, CA, USA). The Brassica 18S rrna probe has been described previously (Park et al., 1993). The expression level was semi-quanti ed using the Phosphoimage analyzer (Fuji photo lm, Tokyo, Japan) and IMAGEGAUGE program provided by the manufacturer. Subcellular localization of the COG1-GFP fusion protein To construct the COG1:GFP fusion, the coding region of the COG1 gene was ampli ed by PCR using primers COG-F (5 0 -CCCGGGG- CGACCCAAAGATTCTCAAGATG-3 0 ) and COG-R (5 0 -GTCGAGA- CAAGATTGACCATCGGTGTA-3 0 ), which resulted in the removal of the termination codon. The PCR product was rst cloned into the pgem-t easy vector (Promega, Madison, WI, USA) to con rm the DNA sequence. The insert was then transferred into the SmaI and SalI sites of the p326gfp-3g vector that contained the GFP gene under the control of the 35S promoter. The resulting construct included an in-frame fusion of the COG1 gene to the 5 0 end of the GFP gene and expressed the COG1:GFP fusion protein under the control of the 35S promoter. The p326gfp-3g vector and the 35S:COG-GFP construct were introduced into onion epidermal cells with a helium biolistic particle delivery system (Bio-Rad, Hercules, CA, USA), as described by Shieh et al. (1993). Expression of the fusion constructs was observed 18 h later by uorescence microscopy using a Zeiss Axioplan uorescence microscope (Jena, Germany). The lters used were XF116 (exciter, 474AF20; dichroic, 500DRLP; emitter, 510AF23) (Omega Inc., Brattleboro, VT, USA) for GFP. The same cells were stained with 1 mg ml 1 of 4 0,6- diamidino-2-phenylindole (Sigma, St. Louis, MO, USA) to identify the nucleus. Acknowledgements This work was supported by the National Research Laboratory grant from the Ministry of Science and Education of Korea (M ), and the work of P.O.L. was partially supported from Korea Research Foundation (2001±050-D00031). We thank Dr Moon Soo Soh for critical reading of the manuscript and Dr In Hwan Hwang for help with localization experiment. The manuscript was edited by Dr Dana Beitner-Johnson at BiomedEditors. References Ahmad, M. and Cashmore, A.R. (1996) The pef mutants of Arabidopsis thaliana de ne lesions early in the phytochrome-signaling pathway. Plant J. 10, 1103±1110. Ballesteros, M.L., Bolle, C., Lois, L.M., Moore, J.M., Vielle-Calzada, J.-P., Grossniklaus, U. and Chua, N.-H. (2001) LAF1, an MYB transcription activator for phytochrome A signaling. Genes Dev. 15, 2613±2625. Barnes, S.A., Nishizawa, N.K., Quaggio, R.B., Whitelam, G.C. and Chua, N.-H. (1996) Far-red light blocks greening of Arabidopsis seedlings via a phytochrome A-mediated change in plastid development. Plant Cell, 8, 601±615. Bolle, C., Koncz, C. and Chua, N.-H. (2000) PAT1, a new member of the GRAS family, is involved in phytochrome A signal transduction. Genes Dev. 14, 1269±1278. Botto, J.F., Sanchez, R.A., Whitelam, G.C. and Casal, J.J. (1996) Phytochrome A mediates the promotion of seed germination by very low uences of light and canopy shade light in Arabidopsis. Plant Physiol. 110, 439±444. Choi, G., Yi, H., Lee, J., Kwon, Y.K., Soh, M.S., Shin, B., Luka, Z., Hahn, T.R. and Song, P.S. (1999) Phytochrome signalling is mediated through nucleoside diphosphate kinase 2. Nature, 401, 610±613. Clough, S.J. and Bent, A.F. (1998) Floral dip: a simpli ed method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735±743. Devlin, P.F., Halliday, K.J., Harberd, N.P. and Whitelam, G.C. (1996) The rosette habit of Arabidopsis thaliana is dependent upon phytochrome action: novel phytochromes control internode elongation and owering time. Plant J. 10, 1127±1134. Devlin, P.F., Patel, S.R. and Whitelam, G.C. (1998) Phytochrome E in uences internode elongation and owering time in Arabidopsis. Plant Cell, 10, 1479±1487. Dieterle, M., Zhou, Y.C., Schafer, E., Funk, M. and Kretsch, T. (2001) EID1, an F-box protein involved in phytochrome A-speci c light signaling. Genes Dev. 15, 939±944. Fairchild, C.D., Schumaker, M.A. and Quail, P.H. (2000) HFR1 encodes an atypical bhlh protein that acts in phytochrome A signal transduction. Genes Dev. 14, 2377±2391. Fankhauser, C., Yeh, K.C., Lagarias, J.C., Zhang, H., Elich, T.D. and Chory, J. (1999) PKS1, a substrate phosphorylated by phytochrome that modulates light signaling in Arabidopsis. Science, 284, 1539±1541. Furuya, M. (1993) Phytochromes: their molecular species, gene families, and functions. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44, 617±645.

10 170 Don Ha Park et al. Genoud, T., Millar, A.J., Nishizawa, N., Kay, S.A., Schafer, E., Nagatani, A. and Chua, N.-H. (1998) An Arabidopsis mutant hypersensitive to red and far-red light signals. Plant Cell, 10, 889±904. Gualberti, G., Papi, M., Bellucci, L., Ricci, I., Bouchez, D., Camilleri, C., Costantino, P. and Vittorioso, P. (2002) Mutations in the Dof zinc nger genes DAG2 and DAG1 in uence with opposite effects the germination of Arabidopsis seeds. Plant Cell, 14, 1253±1263. Guo, H., Mockler, T., Duong, H. and Lin, C. (2001) SUB1, an Arabidopsis Ca 2 -binding protein involved in cryptochrome and phytochrome co-action. Science, 291, 487±490. Hall, T.A. (1999) BIOEDIT: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/ NT. Nucl. Acids Symp. Ser. 41, 95±98. Halliday, K.J., Hudson, M., Ni, M., Qin, M. and Quail, P.H. (1999) poc1: an Arabidopsis mutant perturbed in phytochrome signaling because of a T-DNA insertion in the promoter of PIF3, a gene encoding a phytochrome-interacting bhlh protein. Proc. Natl. Acad. Sci. USA, 96, 5832±5837. Hoecker, U., Tepperman, J.M. and Quail, P.H. (1999) SPA1, a WDrepeat protein speci c to phytochrome A signal transduction. Science, 284, 496±499. Hoecker, U., Xu, Y. and Quail, P.H. (1998) SPA1: a new genetic locus involved in phytochrome A-speci c signal transduction. Plant Cell, 10, 19±33. Hsieh, H.-L., Okamoto, H., Wang, M., Ang, L.-H., Matsui, M., Goodman, H. and Deng, X.W. (2000) FIN219, an auxin-regulated gene, de nes a link between phytochrome A and the downstream regulator COP1 in light control of Arabidopsis development. Genes Dev. 14, 1958±1970. Hudson, M., Ringli, C., Boylan, M.T. and Quail, P.H. (1999) The FAR1 locus encodes a novel nuclear protein speci c to phytochrome A signaling. Genes Dev. 13, 2017±2027. Huq, E., Kang, Y., Halliday, K.J., Qin, M. and Quail, P.H. (2000) SRL1: a new locus speci c to the phyb-signaling pathway in Arabidopsis. Plant J. 23, 461±470. Huq, E. and Quail, P.H. (2002) PIF4, a phytochrome-interacting bhlh factor, functions as a negative regulator of phytochrome B signaling in Arabidopsis. EMBO J. 21, 2441±2450. Kakimoto, T. (1996) CKI1, a histidine kinase homolog implicated in cytokinin signal transduction. Science, 274, 982±985. Kang, H.-G. and Singh, K.B. (2000) Characterization of salicylic acid-responsive Arabidopsis Dof domain proteins: overexpression of OBP3 leads to growth defects. Plant J. 21, 329±339. Kardailsky, I., Shukla, V.K., Ahn, J.H., Dagenais, N., Christensen, S.K., Nguyen, J.T., Chory, J., Harrison, M.J. and Weigel, D. (1999) Activation tagging of the oral inducer FT. Science, 286, 1962±1965. Kim, B.C., Soh, M.S., Hong, S.H., Furuya, M. and Nam, H.G. (1998) Photomorphogenic development of the Arabidopsis shy2-1d mutation and its interaction with phytochromes in darkness. Plant J. 15, 61±68. Kim, B.C., Soh, M.S., Kang, B.J., Furuya, M. and Nam, H.G. (1996) Two dominant photomorphogenic mutations of Arabidopsis thaliana identi ed as suppressor mutations of hy2. Plant J. 9, 441±456. Kunkel, T., Neuhaus, G., Batschauer, A., Chua, N.-H. and Schafer, E. (1996) Functional analysis of yeast-derived phytochrome A and B phycocyanobilin adducts. Plant J. 10, 625±636. Ma, L., Li, J., Qu, L., Hager, J., Chen, Z., Zhao, H. and Deng, X.W. (2001) Light control of Arabidopsis development entails coordinated regulation of genome expression and cellular pathways. Plant Cell, 13, 2589±2607. Mancinelli, A.L. (1990) Interaction between light quality and light quantity in the photoregulation of anthocyanin production. Plant Physiol. 92, 1191±1195. Mena, M., Vicente-Carbajosa, J., Schmidt, R.J. and Carbonero, P. (1998) An endosperm-speci c DOF protein from barley, highly conserved in wheat, binds to and activates transcription from the prolamin-box of a native B-hordein promoter in barley endosperm. Plant J. 16, 53±62. Nagatani, A., Chory, J. and Furuya, M. (1991) Phytochrome B is not detectable in the hy3 mutant of Arabidopsis, which is de cient in responding to end-of-day far-red light treatments. Plant Cell Physiol. 32, 1119±1122. Nagatani, A., Reed, J.W. and Chory, J. (1993) Isolation and initial characterization of Arabidopsis mutants that are de cient in phytochrome A. Plant Physiol. 102, 269±277. Neff, M.M. and Chory, J. (1998) Genetic interactions between phytochrome A, phytochrome B, and cryptochrome 1 during Arabidopsis development. Plant Physiol. 118, 27±35. Neff, M.M., Fankhauser, C. and Chory, J. (2000) Light: an indicator of time and place. Genes Dev. 14, 257±271. Neff, M.M. and Van Volkenburgh, E. (1994) Light-stimulated cotyledon expansion in Arabidopsis seedlings. Plant Physiol. 104, 1027±1032. Ni, M., Tepperman, J.M. and Quail, P.H. (1998) PIF3, a phytochrome-interacting factor necessary for normal photoinduced signal transduction, is a novel basic helix-loop-helix protein. Cell, 95, 657±667. Oyama, T., Shimura, Y. and Okada, K. (1997) The Arabidopsis HY5 gene encodes a bzip protein that regulates stimulus-induced development of root and hypocotyl. Genes Dev. 11, 2983±2995. Papi, M., Sabatini, S., Altamura, M.M., Hennig, L., Schafer, E., Costantino, P. and Vittorioso, P. (2002) Inactivation of the phloem-speci c Dof zinc nger gene DAG1 affects response to light and integrity of the testa of Arabidopsis seeds. Plant Physiol. 128, 411±417. Park, Y.S., Kwak, J.M., Kwon, O.Y., Kim, Y.S., Lee, D.S., Cho, M.J., Lee, H.H. and Nam, H.G. (1993) Generation of expressed sequence tags of random root cdna clones of Brassica napus by single-run partial sequencing. Plant Physiol. 103, 359±370. Parks, B.M. and Quail, P.H. (1993) hy8, a new class of Arabidopsis long hypocotyl mutants de cient in functional phytochrome A. Plant Cell, 5, 39±48. Parks, B.M., Quail, P.H. and Hangarter, R.P. (1996) Phytochrome A regulates red-light induction of phototropic enhancement in Arabidopsis. Plant Physiol. 110, 155±162. Quail, P.H. (2002) Phytochrome photosensory signalling networks. Nat. Rev. Mol. Cell Biol. 3, 85±93. Quail, P.H., Boylan, M.T., Parks, B.M., Short, T.W., Xu, Y. and Wagner, D. (1995) Phytochromes: photosensory perception and signal transduction. Science, 268, 675±680. Reed, J.W., Nagpal, P., Poole, D.S., Furuya, M. and Chory, J. (1993) Mutations in the gene for the red/far-red light receptor phytochrome B alter cell elongation and physiological responses throughout Arabidopsis development. Plant Cell, 5, 147±157. Riechmann, J.L., Heard, J., Martin, G. et al. (2000) Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science, 290, 2105±2110. Robson, P.R.H., Whitelam, G.C. and Smith, H. (1993) Selected components of the shade-avoidance syndrome are displayed in a normal manner in mutants of Arabidopsis thaliana and Brassica rapa de cient in phytochrome B. Plant Physiol. 102, 1179±1184. Schaffer, R., Ramsay, N., Samach, A., Corden, S., Putterill, J., Carre, I.A. and Coupland, G. (1998) The late elongated hypocotyl

11 Role of COG1 in phytochrome signaling 171 mutation of Arabidopsis disrupts circadian rhythms and the photoperiodic control of owering. Cell, 93, 1219±1229. Shieh, M.W., Wessler, S.R. and Raikhel, N.V. (1993) Nuclear targeting of the maize R protein requires two nuclear localization sequences. Plant Physiol. 101, 353±361. Shinomura, T., Nagatani, A., Hanzawa, H., Kubota, M., Watanabe, M. and Furuya, M. (1996) Action spectra for phytochrome A- and B-speci c photoinduction of seed germination in Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA, 93, 8129±8133. Smith, H. (2000) Phytochromes and light signal perception by plants ± an emerging synthesis. Nature, 407, 585±591. Soh, M.S., Hong, S.H., Hanzawa, H., Furuya, M. and Nam, H.G. (1998) Genetic identi cation of FIN2, a far-red light-speci c signaling component of Arabidopsis thaliana. Plant J. 16, 411±419. Soh, M.S., Kim, Y.M., Han, S.J. and Song, P.S. (2000) REP1, a basic helix-loop-helix protein, is required for a branch pathway of phytochrome A signaling in Arabidopsis. Plant Cell, 12, 2061± Somers, D.E., Sharrock, R.A., Tepperman, J.M. and Quail, P.H. (1991) The hy3 long hypocotyl mutant of Arabidopsis is de cient in phytochrome B. Plant Cell, 3, 1263±1274. Spiegelman, J.I., Mindrinos, M.N., Fankhauser, C., Richards, D., Lutes, J., Chory, J. and Oefner, P.J. (2000) Cloning of the Arabidopsis RSF1 gene by using a mapping strategy based on high-density DNA arrays and denaturing high performance liquid chromatography. Plant Cell, 12, 2485±2498. Tepperman, J.M., Zhu, T., Chang, H.-S., Wang, X. and Quail, P.H. (2001) Multiple transcription factor genes are early targets of phytochrome A signaling. Proc. Natl. Acad. Sci. USA, 98, 9437±9442. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G. (1997) The ClustalX windows interface: exible strategies for multiple sequence alignment aided by quality analysis tools. Nucl. Acids Res. 24, 4876±4882. Vicente-Carbajosa, J., Moose, S.P., Parsons, R.L. and Schmidt, R.J. (1997) A maize zinc- nger protein binds the prolamin box in zein gene promoters and interacts with the basic leucine zipper transcriptional activator Opaque2. Proc. Natl. Acad. Sci. USA, 94, 7685±7690. Wagner, D., Hoecker, U. and Quail, P.H. (1997) RED1 is necessary for phytochrome B-mediated red light-speci c signal transduction in Arabidopsis. Plant Cell, 9, 731±743. Wang, Z.Y. and Tobin, E.M. (1998) Constitutive expression of the CIRCADIAN CLOCK ASSOCIATED 1 (CCA1) gene disrupts circadian rhythms and suppresses its own expression. Cell, 93, 1207±1217. Weigel, D., Ahn, J.H., Blazquez, M.A. et al. (2000) Activation tagging in Arabidopsis. Plant Physiol. 122, 1003±1013. Whitelam, G.C. and Devlin, P.F. (1997) Roles of different phytochromes in Arabidopsis photomorphogenesis. Plant Cell Environ. 20, 752±758. Whitelam, G.C., Johnson, E., Peng, J., Carol, P., Anderson, M.L., Cowl, J.S. and Harberd, N.P. (1993) Phytochrome A null mutants of Arabidopsis display a wild-type phenotype in white light. Plant Cell, 5, 757±768. Yanagisawa, S. (1995) A novel DNA-binding domain that may form a single zinc nger motif. Nucl. Acids Res. 23, 3403±3410. Yanagisawa, S. (2002) The Dof family of plant transcription factors. Trends Plant Sci. 7, 555±560. Yanagisawa, S. and Schmidt, R.J. (1999) Diversity and similarity among recognition sequences of Dof transcription factors. Plant J. 17, 209±214. Yanagisawa, S. and Sheen, J. (1998) Involvement of maize Dof zinc nger proteins in tissue-speci c and light-regulated gene expression. Plant Cell, 10, 75±90. Yanovsky, M.J., Casal, J.J. and Luppi, J.P. (1997) The VLF loci, polymorphic between ecotypes Landsberg erecta and Columbia, dissect two branches of phytochrome A signal transduction that correspond to very low uence and high irradiance responses. Plant J. 12, 659±667. EMBL accession number: NM